Advanced Topics in Programming Languages: Transactional Memory at Sun
1:13:55
-
2 years ago
Google Tech Talks
June 14, 2007
ABSTRACT
Concurrent programming is increasingly important and existing programming models are inadequate for this challenge. Transactional memory is the most promising and popular research direction to address these concerns. Software transactional memory is too slow and hardware transactional memory is too complex. Hybrid Transactional Memory runs in existing systems today, but can exploit "best effort" hardware transactional memory support in the future to boost performance. This approach enables and encourages incremental adoption of transactional memory. This talk will describe our recent work on Hybrid Transactional Memory, and some of our other efforts to encourage incremental adoption of transactional memory.
Mark Moir received the B.Sc.(Hons.) degree in Computer Science from Victoria University of Wellington, New Zealand in 1988, and the Ph.D. degree in Computer Science from the University of North Carolina at Chapel Hill, USA in 1996. From August 1996 until June 2000, he was an assistant professor in the Department of Computer Science at the University of Pittsburgh. In June 2000, he joined Sun Microsystems Laboratories, where he is now the Principal Investigator of the Scalable Synchronization Research Group. Dr. Moir's main research interests concern practical and theoretical aspects of concurrent, distributed, and real-time computing. His current research focuses on hardware and software mechanisms for making concurrent programming easier, with a particular focus on transactional memory.Google Tech Talks
June 14, 2007
ABSTRACT
Concurrent programming is increasingly important and existing programming models are inadequ...all »Google Tech Talks
June 14, 2007
ABSTRACT
Concurrent programming is increasingly important and existing programming models are inadequate for this challenge. Transactional memory is the most promising and popular research direction to address these concerns. Software transactional memory is too slow and hardware transactional memory is too complex. Hybrid Transactional Memory runs in existing systems today, but can exploit "best effort" hardware transactional memory support in the future to boost performance. This approach enables and encourages incremental adoption of transactional memory. This talk will describe our recent work on Hybrid Transactional Memory, and some of our other efforts to encourage incremental adoption of transactional memory.
Mark Moir received the B.Sc.(Hons.) degree in Computer Science from Victoria University of Wellington, New Zealand in 1988, and the Ph.D. degree in Computer Science from the University of North Carolina at Chapel Hill, USA in 1996. From August 1996 until June 2000, he was an assistant professor in the Department of Computer Science at the University of Pittsburgh. In June 2000, he joined Sun Microsystems Laboratories, where he is now the Principal Investigator of the Scalable Synchronization Research Group. Dr. Moir's main research interests concern practical and theoretical aspects of concurrent, distributed, and real-time computing. His current research focuses on hardware and software mechanisms for making concurrent programming easier, with a particular focus on transactional memory.«
Download is starting. Save file to your computer. If the download does not start automatically, right-click this link and choose "Save As". How to get videos onto the iPod or PSP.
